To test the hypothesis of whether strenuous physical exercise inhibits neutrophils that can get activated by hypobaric hypoxia, we analyzed the effects of both high altitude and strenuous exercise alone and in combination on potentially cytotoxic functions of granulocytes in healthy volunteers (n = 12 men; average age 27.6 yr; range 24–38 yr). To this end, a field study was prospectively performed with an open-labeled within-subject design comprising three protocols. Protocol I (high altitude) involved a helicopter ascent, overnight stay at 3,196 m, and descent on the following day. Protocol II (physical exercise) involved hiking below an altitude of 2,100 m with repetitive ascents amounting to a total ascent to that of protocol III. Protocol III (combination of physical exercise and high altitude) involved climbing from 1,416 to 3,196 m, stay overnight, and descent on the following day. In protocol I, number of granulocytes did not change, but potentially cytotoxic functions of cells (CD18 expression and superoxide production) were early and significantly upregulated. In protocol II, subjects developed granulocytosis, but functions of cells were inhibited. In protocol III, granulocytosis occurred at higher values than those observed under protocol II. Potentially cytotoxic functions of cells, however, were strongly inhibited again. In conclusion, high altitude alone, even moderate in extent, can activate potentially cytotoxic functions of circulating granulocytes. Strenuous physical exercise strongly inhibits this activation, which may give protection from an otherwise inflammatory injury.

  • mountain
  • polymorphonuclear neutrophil leukocytes
  • purines
  • inflammation

studies on the effects of short-term exercise of moderate intensity showed in general an initial activation of neutrophils as suggested by the release of cytoplasmic enzymes (degranulation) and changes in key effector functions, such as respiratory burst and the phagocytic activity (6, 17, 22). However, the nature of these changes is still unclear, as some studies also show a transient suppression of the respiratory burst and/or phagocytic capacity after exercise even of moderate intensity (14, 20). In contrast to moderate exercise, the reported responses to maximal exercise are more consistent because most polymorphonuclear neutrophils (PMN) functions fall significantly after an episode of exhaustive exercise, and several studies have shown a suppression in the oxidative burst (9, 10, 20, 21). The downregulation of cytotoxic neutrophil functions, especially those that are known to contribute to tissue damage like expression of adhesion molecules (CD11b/CD18) or oxygen radical production, may create a window of opportunity for opportunistic infections to become established but may also help to protect from secondary inflammatory tissue injury associated with strenuous physical exercise.

Together, these observations lend support to the view that limited periods of moderate exercise cause activation, whereas strenuous physical exercise causes functional suppression of circulating granulocytes.

In this context, it is worth noting that many sportsmen undertake physical activities not only at normal altitude but also at higher altitudes as either professionals or vacationers. In fact, the number of recreational climbers or skiers who access high altitudes in the mountainside steadily rises as a result of improvements in touristy infrastructures that allow more and more people to ascend to high altitudes either by hiking or more quickly by funicular railways or helicopters (“heli-skiing”). Interestingly, when healthy volunteers were exposed in decompression chambers intermittently to hypobaric hypoxia to simulate effects of an altitude of 4,500 m, the number of neutrophils transiently increased and the ability of cells to produce superoxide anions (O2) was potentiated (12). Thus mild hypobaric hypoxia is followed by activation of neutrophils, an observation that has been made with the use of normobaric hypoxia in humans as well (26).

Based on these findings, we hypothesized that 1) hypobaric hypoxia as associated with helicopter ascent will stimulate potentially cytotoxic functions of circulating neutrophils and 2) strenuous physical exercise at lower and even at higher altitude will inhibit neutrophil activation.

To test our hypothesis, we performed a prospective, controlled, open-labeled field study with a within-subject-design, in which the effects of both high altitude and strenuous exercise were analyzed individually and in combination with respect to parameters of tissue toxic neutrophil functions in healthy volunteers.

The results of this study show that high altitude alone, even moderate in extent, can indeed activate potentially cytotoxic functions of circulating granulocytes. Strenuous physical exercise, however, strongly inhibits this activation, which may give protection from otherwise inflammatory tissue injury.


Study Protocols

After approval by the local ethical committee and informed consent was obtained from the subjects, 12 healthy, young mountaineers (average age 27.6 yr; range 24–38 yr) were included in the study, which took place in Italy in the South Tyrolean Alps at the Becherhaus summit (3,196 m). The base camp was located in Ridnaun Valley at 1,416 m. The volunteers had no congenital or acquired heart defects and underwent medical investigation before the study. All study participants received supervision by the same experienced medical and laboratory staff during the entire study and were not allowed to take self-prescribed drugs before and during the study. The study included three different protocols. A time interval of >6 wk was kept between each study protocol. The range of daytime temperatures did not vary more than ±5°C throughout the whole study. Protocol I involved an ascent to high altitude by helicopter, overnight stay on the summit, and flight back to the base camp in the valley on the following day 24 h after ascent (flight time: 8 min each way). Subjects stayed at rest during the whole observation period. Protocol II involved strenuous physical exercise at a lower altitude that did not exceeding 2,100 m above sea level. To achieve a comparable degree of exercise as under protocol III, volunteers climbed to an altitude <2,100 m four times within a total hiking time of 4–4.5 h (the difference in altitude being 1,644 m altogether). Overnight stay was at the base camp (1,416 m). Protocol III involved combined effects of high altitude and strenuous physical exercise by climbing from the base camp at 1,416 m to the Becherhaus summit at 3,196 m above sea level (1,780-m difference in altitude; hiking time: 3–4 h), recovery and overnight stay on summit hut, and descent on the following day 24 h after ascent. Medical examination and blood sampling from the antecubital vein were carried out four times on each subject: before start (T0), at the summit lodge 60–90 min on arrival (T1) and after the overnight stay (T2), as well as 60–90 min after return to the base camp on the following day (T3). For protocol II, medical examination and blood collection were always performed in the base camp. All subjects underwent a complete physical examination, and noninvasive systemic blood pressure, respiratory frequency, pulse rate, oxygen saturation by pulse oximetry (Siemens SC6002, Siemens AG, Erlangen, Germany) were measured. To exclude development of acute mountain sickness (AMS), subjects were monitored by the Lake Louise questionnaire. The questionnaire is focused on symptoms related to AMS (headache, vegetative symptoms, fatigue), and each symptom was scored from 0 (none) to 3 (distinct). The individual score of each subject was calculated by adding the scores of each parameter. In accordance with others (15, 23), a cumulative score of three or more points was considered as AMS of moderate severity and a score of more than six points as AMS of severe degree.

Blood Assays

Cell counts were performed on EDTA-anticoagulated blood samples (Coulter STKS, Coulter Electronics, Luton, UK). Additionally, EDTA-anticoagulated blood specimens were centrifuged at 600 g for 5 min in a cooled centrifuge at 2°C and stored at −80°C (dry ice) for determination of IL-6 (TINA-Quant, Roche Diagnostics, Mannheim, Germany). Plasma lactate was determined from 2 ml of blood, which were collected in tubes containing 5 mg of fluoride EDTA. After centrifugation (5 min, 600 g), plasma was stored at 0°C until determination of the lactate concentration by the lactate oxidase assay (COBAS Integra, Roche Diagnostics, Basel, Switzerland).

Functional analyses.

The expression of β2-integrins (CD18) on PMN leukocytes was measured by flow cytometery (FACScan, Becton Dickenson) in heparinized blood following incubation of cells with fluorescence-labeled monoclonal anti-CD18 antibody. Data are expressed in relative fluoresence units, as reported previously (29). The capability of granulocytes to produce O2 under resting and stimulated conditions was assessed in diluted whole blood samples by spectrophotometric determination of the reduction of cytochrome C (7). Briefly, immediately after blood sampling, heparinized whole blood was incubated (15 min, 37°C) in tubes prefilled with Hanks’ buffered salt solution containing cytochrome c (0.625 mg/ml), cytochalasin B (2.5 μg/ml), superoxide dismutase (0 or 50 U/ml) in the absence and presence of the chemotactic tripeptide N-formy-methionyl-phenyl-alanine (fMLP; 10−6 M). Thereafter, cell-free supernatants were collected and stored at −20°C until determination in the laboratory within 2 wk. From the differences calculated between the absorbances (spectrophotometry, 540 nm) of samples incubated with or without superoxide dismutase, the amount of O2 was derived (nmol/1 × 106 PMN/15 min).

Plasma Catecholamines

Epinephrine and norepinephrine concentrations were determined in cell-free supernatants from blood that was centrifuged immediately after collection by venipuncture. Thereafter, plasma samples were immediately frozen and stored at −80°C until determination of epinephrine and norepinephrine concentrations by high-performance liquid chromatography (Chromosytems, Martinsried, Germany).

Plasma Purines

To get insight into the effects of possible tissue hypoxia, plasma concentrations of adenosine and its degradation products inosine and hypoxanthine were measured. These purines are well documented to be sensitive indicators of enhanced degradation of adenine nucleotides (ATP, ADP, AMP) when oxygen demand exceeds oxygen supply (28). To determine plasma concentrations of adenosine, inosine, and hypoxanthine, blood samples (2 ml) were collected and filled into ice-cooled syringes containing the following reagents: physiological saline, 2 × 10−4 M dipyridamole, 2 × 10−5 M erythro-9-(2-hydroxy-3-nonyl)adenine, 2 × 10−2 M EDTA, 2 × 10−2 M EGTA, and 2 × 10−2 M dl-α-glycerophosphate at a pH of 7.4. These compounds prevent removal of adenosine or its metabolites from plasma by inhibition of adenosine reuptake and degradation. Inhibition of ectonucleotidases and nonspecific phosphatases additionally stops the formation of purines from adenine nucleotides. The samples were transferred into tubes that were spun down (5 min, 600 g), and supernatants were preserved in perchloric acid (final 35%) at −80°C. Inosine and adenosine concentrations (in nM) were analyzed by reversed-phase high-performance liquid chromatography technique (11). The purine base hypoxanthine was also determined by reversed-phase high-performance liquid chromatography using a modified method published previously (18).

Adenosine and O2 Production

Because the nucleoside adenosine has previously been shown to strongly inhibit the numerical upregulation of β2-integrins (29) and the production of oxygen radicals of human polymorphonuclear leukocytes (27), we also characterized the effects of increasing concentrations of adenosine on the production of O2. Granulocytes were separated from whole blood of healthy volunteers by a modified method described by Boyum (3) and Chouker et al. (7). Briefly, heparinized blood was mixed with an equal amount of dextran 60, and erythrocyte was allowed to sedimentate for 45 min. The leukocyte-enriched supernatant was removed and centrifuged on a discontinuous Ficoll-Histopaque gradient (density: 1,077 mg/ml). PMN leukocytes were harvested and washed with Hanks’ buffered salt solution, and contaminating erythrocytes were lysed thereafter. The capacity of a standardized number of separated PMN leukocytes to produce O2 was determined under the same conditions of incubation as described above but in the absence or presence of increasing concentrations of the metabolically stable adenosine receptor nonspecific agonist 2-Cl-adenosine (0–10,000 nM).

Statistical Analysis

On rejection of the null hypothesis (data are not normally distributed) by one-sample Kolmogorov-Smirnov test, within-group and between-group comparisons were performed by repeated-measures ANOVA or ANOVA only, followed by paired and unpaired t-test, respectively. Two-tailed level of significance was set at P < 0.05 and corrected for multiple comparisons by Bonferoni. Data are presented as means ± SE. Bivariate correlation according to Pearson was calculated between plasma concentrations of adenosine and the fMLP-stimulated production of O2. All statistical analyses were performed by SPSS program 10.0 (SPSS, Chicago, IL).


Baseline measurements (T0) were not significantly different between Protocols I, II, and III and were all in the physiological range (Tables 14).

View this table:
Table 1.

Volunteers’ characteristics, cardiovascular, and respiratory changes during 3 different experimental protocols

Protocol I: Ascent by Helicopter Flight to High Altitude

After ascent by helicopter flight, blood oxygen saturation significantly decreased compared with values measured at the base camp (Table 1; SpO2 T1 vs. T0). None of the 12 individuals developed symptoms of high-altitude pulmonary edema (HAPE) or AMS (AMS scores of <3 points; data not shown). Leukocyte, granulocyte, or lymphocyte counts did not change; however, concentration of hemoglobin significantly increased (> 1 g/l) and remained elevated after return of subjects to the valley (Table 2). Although number of granulocytes did not change, functions of these cells known to contribute to inflammatory tissue damage were significantly upregulated (Table 3). Accordingly, expression of CD18 adhesion molecules on circulating PMN leukocytes was increased by >20%, and spontaneous production of O2 was sixfold higher than the values determined in the valley (T0). In contrast to the early activation of circulating granulocytes, plasma concentrations of IL-6 did not change. Plasma concentrations of epinephrine (Table 4) could be determined in only five volunteers before helicopter ascent for technical reasons so that intragroup comparison failed to detect any changes. However, compared with the time course of epinephrine concentrations determined during protocols II and III, a significant elevation was evident. Plasma concentrations of norepinephrine started to increase on arrival on the summit, doubled after overnight stay (Table 4; T2), and were still significantly higher compared with baseline values after return to the base camp (Table 4; T3). Plasma concentrations of purines did not change, whereas those of lactate paralleled the course of concentrations of epinephrine and norepinephrine.

View this table:
Table 2.

Concentration of circulating white blood cells and hemoglobin

View this table:
Table 3.

β2-Integrin expression and extracellular superoxide anion release by granulocytes and IL-6 plasma concentrations

View this table:
Table 4.

Concentration of catecholamines, purines, and lactate

Protocol II: Strenuous Physical Exercise Below 2,100 m

Exercise by hiking below an altitude of 2,100 m had no effect on partial blood oxygen saturation. White blood cell counts showed a pronounced leukocytosis due to granulocytosis and lymphocytosis. There were no changes in the concentration of hemoglobin (Table 2). Although the number of granulocytes significantly increased, potentially cytotoxic functions of these cells were not activated. In contrast, stimulation of O2 production with the chemotactic tripeptide fMLP was significantly inhibited early after completion of physical exercise (Table 3; T0 vs. T1) but normalized thereafter. Plasma concentrations of IL-6, epinephrine, and norepinephrine followed the time course of granulocytosis, reaching peak values at T1. Plasma concentrations of purines peaked with adenosine, inosine, and hypoxanthine also at T1, changes that were reflected by plasma lactate.

Protocol III: Climbing to High Altitude

Strenuous exercise by hiking to the summit (3,196 m) was followed by a significant decrease in partial blood oxygen saturation but was not different from values observed after helicopter ascent (Table 1). None of the individuals developed symptoms of HAPE or AMS (AMS scores of <3 points; data not shown). White blood cell counts showed a pronounced leukocytosis due to a massive granulocytosis. Lymphocyte counts did not change nor did the concentration of hemoglobin (Table 2). Although the number of granulocytes strongly increased, granulocytes were not activated. By contrast, on arrival at the summit (T1), production of O2 stimulated by the chemotactic tripeptide fMLP was significantly suppressed (Table 3). Thereafter, granulocytosis declined in parallel with reversal of inhibition of the fMLP-stimulated O2 production (T2). After descent (T3), granulocytosis and inhibition of stimulated O2 production occurred again but in a less-pronounced way. Plasma concentrations of IL-6, epinephrine, and norepinephrine followed the time course of granulocytosis, exhibiting increases and decreases on exercise and rest, respectively. Purine plasma concentrations followed the same pattern as observed for changes of the number of granulocytes, IL-6, and catecholamines. Lactate concentrations were highest at T1.

Relationship Between Plasma Concentrations of Adenosine and Inhibition of O2 Production

The nucleoside adenosine is not only a sensitive marker of tissue hypoxia but is also a well known inhibitor of various functions of phagocytes and lymphocytes (28). Because in the present study O2 production elicited by fMLP was always inhibited early after strenuous physical exercise, i.e., circumstances under which plasma concentrations of adenosine are expected to be high, we 1) tested whether our O2 assay was able to also detect inhibitory effects of adenosine in vitro and 2) looked for a possible relationship between the in vivo plasma concentrations of adenosine and the ex vivo determined capacity of granulocyte to produce O2.

As can be seen in Fig. 1A, 2-Cl-adenosine, a metabolically stable analog with binding affinities comparable to the native molecule adenosine, dose-dependently inhibited the fMLP-stimulated O2 production in the assay used. In agreement with the action of adenosine at an adenosine A2A receptor site, the dose-response curve was of sigmoidal shape, reaching maximal inhibitory effects at 1 μM. The IC50 was ∼30 nM, which is well in the range of physiological concentrations. Based on the finding that our assay may well detect alterations in the O2 production of granulocytes induced by adenosine, we looked for a negative correlation between the in vivo plasma concentrations of adenosine and the ex vivo determined fMLP-stimulated O2 production in the 12 healthy volunteers. Due to technical problems, concentrations of adenosine could not be determined in plasma samples taken on the summit hut but could during protocol II, i.e., during hiking in the valley. There was an inverse relationship between the individual adenosine plasma concentrations and the ability of granulocytes to produce O2 in response to fMLP (r = −0.478; P < 0.05) and the IC50 of adenosine was found to be 80 nM (Fig. 1B).

Fig. 1.

In vitro and in vivo effect of adenosine on the respiratory burst of polymorphonuclear neutrophil leukocytes (PMNL) after stimulation with N-formy-methionyl-phenyl-alanine (fMLP). A: in vitro effects of 2-chloro-adenosine (2-Cl-Adenosine) on the fMLP-induced superoxide anion (O2) production (O2 × 106 PMNL × 15 min−1, linear scale). PMNL were separated and incubated with fMLP and increasing concentrations of 2-Cl-adenosine. The IC50 of adenosine was ∼30 nM. Values are means ± SE (error bars); n = 6 subjects. B: every point represents the rate of O2 production (O2 × 106 PMNL × 15 min−1, linear scale) and the respective adenosine plasma concentration (nM, log-scale) for each subject before start, at the summit lodge 60–90 min on arrival, and after the overnight stay during protocol II. Data were analyzed by linear regression and are described by the equation y = −0.375x + 49.19, which shows an inverse relationship. IC50 of adenosine was 80 nM.


The present study was designed to investigate in healthy volunteers, who were not susceptible to AMS, the role of high altitude-associated hypoxia and strenuous physical exercise both alone and in combination on cardiopulmonary, immunologic, hormonal, and metabolic parameters. Specifically, we looked for changes in the number and potentially cytotoxic functions of circulating granulocytes during passive and active ascent. The major finding is that acute exposure to mild hypoxia by passive helicopter ascent resulted in upregulation of potentially cytotoxic functions of granulocytes, whereas active ascent by hiking to the top of the mountain mitigated the activation of these cells.

Protocol I: Passive Ascent by Helicopter Flight

On transport of individuals to the summit by helicopter ascent, numerical upregulation of β2-integrins and several-fold enhanced spontaneous release of O2 became evident within 60–90 min. Concentrations of IL-6 did not change, excluding systemic inflammation as a cause for activation of granulocytes. Unchanged IL-6 concentration under hypoxic conditions was an unexpected finding because others had reported an increase of IL-6 in response of humans to acute and chronic hypoxia (16). However, one group found no increases of IL-6 at 1 or 2 days at 4,000 m (19) or a more gradual increase to 4 days after altitude exposure at 4,350 m (13). Thus the length of the time period and the height of the altitude to which individuals were exposed in our study may not have been sufficient to elevate IL-6 plasma levels.

To our knowledge, only two studies have investigated the effect of hypobaric or normobaric hypoxic exposure of humans on granulocytes functions ex vivo. In agreement with the present findings, repeated hypobaric hypoxia increased oxygen radical production (12), a result that was also found even after a single period of hypoxia (26). However, in these investigations, effects of hypoxia were assessed after reoxygenation and after additional activation of cells by soluble or particulate stimuli, resembling more the effects of hypoxia reoxygenation on stimulated granulocyte activation. Nonetheless, hypobaric hypoxia and reoxygenation-induced granulocyte activation was considered relevant for the pathogenesis of altitude-associated disease (12). In this context, it is tempting to speculate on the fact that the significant rise in the blood concentration of hemoglobin by >1 g/dl could reflect an increase in microvascular permeability followed by a fluid shift out of the blood. This all the more as the time for erythropoiesis was much too short, and body weight did not change, indicating no dehydration (Table 1).

Protocol II: Hiking Below 2,100 m

Physical exercise in the valley caused an impressive leukocytosis mainly due to the rapid onset of granulocytosis in combination with lymphocytosis. Differential smears revealed an increase mostly of mature granulocytes. Granulocytes, however, were not in an activated state because neither numerical expression of β2-integrins nor the spontaneous production rate of O2 increased. By contrast, O2 production stimulated by the chemotactic tripeptide fMLP was inhibited 60–90 min after exercise (T1). Because plasma concentrations of IL-6 were increased, a finding explained by others as the consequence of a (micro) trauma-induced inflammatory response (5), the observed lack of granulocyte activation could indicate the influence of some humoral suppressive factors as also suggested by others (9, 10, 14, 21).

Protocol III: Climbing to 3,196 m

Because passive ascent to the summit by helicopter flight was followed by neutrophil activation, whereas hiking in the valley caused functional depression of granulocytes, the question was what combined effects would look like. In further support of exercise being associated with the release of endogenous inhibitory agents, parameters of granulocyte activation were maximally suppressed again on arrival at the summit. There was also no indirect evidence for an inflammation-associated increase in vascular permeability, as suggested by unchanged hemoglobin concentrations despite a tendency toward lower body weights.

Novelty of Findings and Potential Mechanisms Involved in Exercise-Induced Immunosuppression

Although experiments with high-altitude decompression chambers had already shown stimulation of oxidative burst activity of human granulocytes (12, 26), this is to our knowledge the first time that activation of granulocytes by passive ascent of humans to higher altitude leading to mild hypoxia has been demonstrated. Moreover, this ascent-induced in vivo activation of circulating granulocytes could be attenuated by acute strenuous physical exercise. With respect to the latter, abundant studies already demonstrated mostly suppression of oxidative neutrophil functions by even moderate (14, 20) and exhaustive physical stress (9, 10, 21). Thus we hypothesized that inhibition of high-altitude hypoxia induced neutrophil activation by strenuous exercise, but this prediction had not been tested in a study before. Although the full mechanisms of these responses could not be elucidated by our study, there is some descriptive evidence for the neuroendocrine response to be involved in the downregulation of neutrophil function by strenuous physical stress. Granulocyte counts reached peak values on completion of maximum physical work (T1; protocols II and III), most probably as a consequence of demargination mediated by catecholamines. Indeed, granulocytosis has been known for more than one century to occur in response to muscle activity, which can be mimicked by the administration of epinephrine (2). Endogenous secretion of epinephrine also appears to trigger the efflux and trafficking of granulocytes, e.g., from the spleen (33), and from the marginal pool in part by increasing blood flow velocities during physical exercise (2). Accordingly, time courses of exercise-associated granulocytosis and plasma concentrations of epinephrine followed almost the same kinetic. Because suppression of fMLP-stimulated O2 production was maximal when epinephrine and norepinephrine plasma concentrations together achieved the highest values, both catecholamines are likely candidates to inhibit activation of phagocytes. In further support of a rise of catecholamines to physiologically regulatory concentrations is the increase in plasma lactate concentrations, which may reflect stimulation of glycolysis by epinephrine. Moreover, inhibition by epinephrine and norepinephrine of radical production of human granulocytes has also been described in vitro (32). However, this does not rule out that some other or even more efficient inhibitory mechanisms may have taken effect. In this regard, it is interesting to note that plasma concentrations of purines, especially those of adenosine, were inversely correlated to the ability of circulating granulocytes to produce O2 when stimulated ex vivo by fMLP (Fig. 1). In addition, adenosine dose-dependently inhibited fMLP-stimulated O2 production, showing comparable pronounced inhibition for the range of plasma concentrations determined in vivo. Moreover, adenosine, activating adenosine A2A receptors, was not only reported to inhibit neutrophil respiratory burst activity (27) and CD18 expression (29) but also to act in strong synergy with catecholamines (1). Both A2A adenosine receptor- and β-receptor-dependent pathways have the adenylyl cyclase as a common target, the stimulation of which is well known to increase intracellular cAMP, leading to the pronounced inhibition of neutrophil functions. Other than β-adrenergic catecholamines and adenosine, it was also discussed that prostaglandins potently activate the adenylyl-cyclase system and thereby cause paralysis of the innate immune response. However, inhibition of cyclooxygenase by oral indomethacin did not prevent suppression of the natural killer cell-mediated cytotoxicity caused by 1 h of high-intensity running (4).

Thus it appears to be a valid suggestion that adenosine and β-adrenergic catecholamines are likely candidates that act synergistically in the downregulation of neutrophil functions to provide protection from exercise-induced inflammatory tissue damage that otherwise might be amplified by cytotoxic granulocytes.

Relevance of Findings for Pathogenesis of HAPE

Our observations might also provide a possible explanation of why usually no severe inflammatory tissue damage occurs despite development of pronounced leukocytosis after strenuous physical excercise. These findings are also in line with the lack of activated granulocytes and biochemical signs of lung inflammation in the very early phase of HAPE (25). On the other hand, it is tempting to speculate for a possible failure of the exercise-induced downregulation of cytotoxic properties of granulocytes as a contributing factor in the amplification of inflammatory processes during the more progressed phase of HAPE (24). Thus it would be interesting to make similar measurements in individuals susceptible to altitude illnesses (AMS and HAPE) to test the hypothesis that susceptible subjects have less inflammatory inhibition. The results of this study also necessitate reconsidering the pathophysiological consequences of prolonged exposure to high altitude alone, especially as an increasing number of tourists are lifted to alpine recreational areas by funiculars, helicopters, and cars. Finally, thousands of passengers are exposed day by day to mild hypobaric conditions during commercial airline flights despite equipment of airplanes with compression chambers with air pressure equal to an altitude of up to 2,500 m (8). However, these issues warrant further investigation.

Study Limitations

It might have been preferable to conduct the protocols in an altitude chamber so as to clearly separate the pure effects of altitude from temperature or humidity changes and clearly separate the effects of rest from exercise and to adjust atmospheric pressure to that of sea level rather than to 1,416 m of the base camp. With respect to the degree of physical exercise, it is also worth mentioning that, despite a comparable elevation gain (meters of altitude) during protocols II and III, a continuous hike to the summit may not be fully equivalent to a tour with repetitive hiking up and down. Moreover, since it has to be recognized that, in a field study, measurements of variables are taken at isolated points in time and subject to environmental influences that cannot be controlled, the ability of this study to precisely describe cause and effect between measured variables is limited. Given the complexity and redundancy of the immune system and the neuroendocrine and metabolic systems, it is also almost impossible to establish a direct cause-and-effect relationship between parameters of neutrophil activation and its possible regulation by catecholamines and adenosine. Rather, the observed negative correlation between plasma concentrations of adenosine and the ex vivo-determined capability of granulocytes to produce O2 suggests more focused studies to evaluate cause-and-effect relationships in the regulation of cytotoxic neutrophil functions by physical exercise.

Finally, since none of the subjects were ill with high-altitude illness at the moderate altitude of 3,196 m, failure of the suggested regulation of neutrophil functions by purines and catecholamines could not be tested as a possible cause of HAPE at higher altitude.

On the other hand, the results of this field study performed in the same 12 volunteers showed that high altitude alone, even moderate in extent, can activate potentially cytotoxic functions of circulating granulocytes. Strenuous exercise strongly inhibits this activation, which may protect from otherwise inflammatory injury.


This field study was supported by a research grant from the South Tyrolean Department of Health and by Siemens Medical Solutions, Munich, Germany, the Clinic of Anaesthesiology, and Department of Intensive Care Medicine of Brixen (Italy) and Munich (Germany), respectively.


The authors thank the volunteers of the South Tyrolean Mountain Rescue Service who generously spent their time and participated with enthusiasm. We also thank Dr. G. Andergassen, Dr. G. Rammlmair, S. Kofler, Dr. M. Choukèr, M. Hoelzl, O. Zorzi, Dr. C. Moser, Dr. M. Niklas, Dr. J. Abicht, S. Schröpfer, and Dr. M. Vogeser for continuous help during the study and B. Lobstein for the revision of the maunscript.


  • * A. Choukèr and F. Demetz contributed equally to this work.

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract